The ability of the dark-adapted visual system to count photon absorptions has been known for many years; however, the biophysical mechanisms which make photon detection and counting possible are not understood. Much of rod vision occurs at light levels where photon absorptions occur rarely; thus, failure of these biophysical processes will severely impair rod vision. The reliability of photon counting means that the transduction process in the rod outer segment must produce distinguishable responses to absorption of O, 1, or 2 photons, and that these signals are reliably transmitted from the rod to the rest of the visual system. Current understanding of signal transduction and synaptic transmission cannot account for this level of performance. Experiments proposed here will determine the mechanisms mediating reliable transduction of single photon absorptions and reliable transmission of these responses to second order cells. Studies of phototransduction will determine how each absorbed photon produces a nearly identical current response. This reproducibility is essential for reliable photon counting, but is surprising given the expected statistical fluctuations in the small number of molecules involved, particularly in the shutoff of photoexcited rhodopsin. Elements of the transduction cascade will be selectively altered to determine if they are necessary for reproducible single photon responses. Studies of synaptic transmission will determine how single photon responses are reliably transmitted to second order cells in the retina, and how pre-synaptic mechanisms separate light signals from photoreceptor noise. The single photon signal traversing the synapse and the noise generated in synaptic transmission will be studied by monitoring the post-synaptic current in a bipolar cell while controlling the rod voltage. Signal transfer to bipolar cells also separates light signals from photoreceptor noise. A likely mechanism mediating this synaptic filtering is the control of the Ca2+ concentration in the synaptic terminal. To test this idea, the kinetics of Ca2+ changes in the terminal will be measured with fluorescent indicators and compared to the kinetics of signal transfer. A further test will be made by determining how the kinetics of signal transfer are altered when the buffering capacity of the terminal is changed.